Erratum (01 May 1970)

The factor exp [--iQ30 -- 0)z] appearing in equations (45), (53) and (54) should be replaced by this correct phase factor. All other equations remain valid. The error has the effect of letting the radiation loss of a linear taper shown in Fig. 8 appear too high. The point at L = 0 remains the same but the loss at L/dx = 40 is reduced from 0.036 to 0.014. All other curves remain correct. I am grateful to Mr. A. L. Jones of the IBM Corporation for pointing this error out to me. (D.M.) 919